Training Data With Azure ML

Koos van Strien discusses training data sets and cross-validating results:

When choosing a train and testset, you’ll implicitly introduce a new bias: it could be that the model you just trained predicts well for this particular testset, when trained for this particular trainset. To reduce this bias, you could “cross-validate” your results.

Cross-validation (often abbreviated as just “cv”) splits the dataset into n folds. Each fold is used once as a testset, using all other folds together as a training set. So in our pizza example with 100 records, with 5 folds we will have 5 test runs:

This isn’t Azure ML-specific, and is good reading.

Related Posts

Picking An Azure SQL Database Tier

Esat Erkec has various methods you can use to figure out your Azure SQL Database tier: When we are beginning to think of migrating our on-premises databases to Azure SQL, we have to decide on a proper purchase model, a service tier, and a performance level. Before starting the Azure SQL migration process, we have […]

Read More

Gartner’s Cloud IaaS Magic Quadrant Changes

Bruno Aziza analyzes Gartner’s Magic Quadrant for Cloud Infrastructure as a Service offerings: The first and most drastic change that occurred over the last year is the number of players that Gartner decided to highlight in its report: the number of vendors went from 14 to just 6 this year.   Why is that?! Have the big become bigger […]

Read More

Categories

August 2016
MTWTFSS
« Jul Sep »
1234567
891011121314
15161718192021
22232425262728
293031